use "citation_gap.dta", clear


//requires the installation of cibar 

//make graphs black and white for publication
set scheme s1mono

//figure one (probability reviewer cites a woman)

cibar revfemcite, over1(revfem) over2(treated) level(90) graphopts(ytitle("P(Reviewer cites a woman)") yscale(range(0 .5)) ylabel(0(0.1).5) text(0.01 1.0 "n=103", place(c)) text(0.01 2 "n=60", place(c)) text(0.01 3.7 "n=202", place(c)) text(0.01 4.7 "n=123", place(c)))




// figure two (number of women cited given a woman is cited)
cibar numfemcites if revfemcites==1, over1(revfem) over2(treated) level(90) graphopts(ytitle("Number of women cited | woman cited") yscale(range(0 2.5)) ylabel(0(0.5)2.5) text(0.05 1.0 "n=103", place(c)) text(0.05 2 "n=60", place(c)) text(0.05 3.7 "n=202", place(c)) text(0.05 4.7 "n=123", place(c)))


//figure three (reviewer self-cites)

cibar revselfcite, over1(revfem) over2(treated) level(90) graphopts(ytitle("P(Reviewer self-cites)") yscale(range(0 .2)) ylabel(0(0.05).2) text(0.005 1.0 "n=103", place(c)) text(0.005 2 "n=60", place(c)) text(0.005 3.7 "n=202", place(c)) text(0.005 4.7 "n=123", place(c)))


//materials for online appendix

by treated, sort : summarize numfemcites nummalecites

by treated, sort: summarize revfem grad asst assoc full rankofreviewer

logit revselfcite revfem#treated rankofreviewer
